Coronin 1B coordinates Arp2/3 complex and cofilin activities at the leading edge

Abstract

Actin filament formation and turnover within the treadmilling actin filament array at the leading edge of migrating cells are interdependent and coupled, but the mechanisms coordinating these two activities are not understood. We report that Coronin 1B interacts simultaneously with Arp2/3 complex and Slingshot (SSH1L) phosphatase, two regulators of actin filament formation and turnover, respectively. Coronin 1B inhibits filament nucleation by Arp2/3 complex and this inhibition is attenuated by phosphorylation of Coronin 1B at Serine 2, a site targeted by SSH1L. Coronin 1B also directs SSH1L to lamellipodia where SSH1L likely regulates Cofilin activity via dephosphorylation. Accordingly, depleting Coronin 1B increases phospho-Cofilin levels, and alters lamellipodial dynamics and actin filament architecture at the leading edge. We conclude that Coronin 1B's coordination of filament formation by Arp2/3 complex and filament turnover by Cofilin is required for effective lamellipodial protrusion and cell migration.

Figure 1. Depletion of Coronin 1B slows whole cell migration and alters lamellipodial dynamics

A- Target sequence of an shRNA designed to knock down expression of mouse or rat, but not human, Coronin 1B. B- HEK293 cells were co-transfected with the Coronin 1B shRNA and GFP-tagged human or mouse Coronin 1B. Cell lysates were immunoblotted with anti-GFP to verify the efficiency and species specificity of knockdown. p34Arc protein was detected as a loading control. C- NIH3T3 fibroblasts were transfected with Coronin 1B shRNA (indicated by GFP fluorescence) for 72 hours, and immunostained for Coronin 1B (red) to verify knockdown of the endogenous gene. D- Mean cell speeds of Rat2 fibroblasts infected with lentivirus to express Coronin 1B shRNA (KD-1B, without or with (rescue) expression of GFP-tagged human Coronin 1B) or control shRNA (NS). Error bars represent the 95% confidence intervals. Newman-Keuls multiple comparison test was used after one-way ANOVA to generate the P values (P<0.001, ***). E- Method for kymography analysis. Minimal intensity projection of a 300-frame 1-second-interval movie was presented on the right. Pixel intensities along a 1-pixel wide line (blue) were used to generate the kymograph presented in the blue box; a magnified region (outlined in green) is displayed on the right. Red dashed lines indicate the parameters for one protrusion. D is protrusion distance, P is protrusion persistence, and tanα is protrusion rate. F- Protrusion parameters of Rat2 cells infected with lentivirus expressing the Coronin 1B shRNA without or with the rescue construct for expression of GFP–tagged human Coronin 1B (rescue). Sample kymographs corresponding to each treatment are shown above each bar; red lines indicate persistence time for each protrusion. The mean value for each parameters is presented; error bars indicate the 95% confidence intervals. Newman-Keuls multiple comparison test was used after one-way ANOVA to generate the P values (***P<0.001 when compared to Rat2 cells).

Figure 2. Depletion of Coronin 1B slows retrograde actin flow, influences barbed end distribution and filament architecture at the leading edge

GFP-actin in Rat2 cells infected with Lentivirus expressing Coronin 1B (KD-1B) or control (NS) shRNA and GFP-actin was imaged for at 1 sec intervals for 2 min. A- Three representative kymographs showing retrograde actin flow are presented for each condition. Red bar = 1.14 µm; white bar = 30 sec. B- Average actin retrograde flow rate in KD-1B and NS expressing cells (3 measurements/cell, n = 40 cells) for each condition are presented as box and whisker plots (Dot = mean, middle line = median, top & bottom of box = 75% and 25%, whiskers = full data range). Unpaired student t-test indicates a signification difference between samples (P<0.0001). C- KD-1B- and NS-expressing Rat2 cells were subjected to the barbed end assay using Alexa Fluor 568-labeled G-actin. Two cells for each condition are shown. Scale bar = 10 µm. D- Quantification of free barbed ends in KD-1B and NS cells. Pixel intensities of Alexa-568-actin around the leading edge were plotted as described in Fig. S1. The region encompassing the top 50% of the barbed end signal is defined as the width of the barbed end zone (see Fig. S2). Width of this zone in KD-1B and NS cells is presented as a box and whisker plot. Unpaired student t test indicates a significant difference between the samples (P=0.0165). E- The normalized ratio of barbed ends (Alexa-568 actin signal) to total actin (AlexaFluor647-phalloidin signal) was determined as a function of distance from the cell edge (see Fig. S2). Data were from cells shown in D and presented as mean with error bars indicating standard errors of the mean. Paired student t-test indicates a significant difference in barbed end density relative to F-actin between Coronin 1B-depleted and control cells (P=0.013 for the region 0.3µm - 1.4µm from the edge). F- Platinum replica electron micrographs of lamellipodia in Coronin 1B-depleted and control Rat2 fibroblasts. Expanded views of each cell is presented as an inset. Branched actin filaments are pseudo colored with yellow and green. Scale bar = 500 nm.

Figure 3. Coronin 1B inhibits actin nucleation by Arp2/3 complex in vitro

A- Plotted are actin polymer concentrations versus time in reactions containing 1.5 µM actin (5% pyrene labeled), 20 nM Arp2/3 complex, 10 nM GST-VCA, and Coronin 1B as indicated in figure. The curves labeled actin (yellow) and Arp2/3+actin (tan) contained neither Coronin 1B nor GST-VCA. B- Phospho-Coronin 1B is less potent at inhibiting Arp2/3 complex. Plotted are the maximal rates of actin polymerization at varying Coronin 1B concentrations. WT, wild-type Coronin 1B (no detectable phosphorylation); p-WT, wild type Coronin 1B purified from PMA-stimulated cells (~75% phosphorylated); S2D, a mutant Coronin 1B with a phospho-mimetic aspartate substitution at Ser2. Maximal assembly rates were calculated from a linear fit to the rate of actin polymerization where 0.5–1.0 µM F-actin formed during each reaction. Data are derived from experiments presented in Fig. S7. C- Coronin 1B binds Arp2/3 complex directly. His-tagged Coronin 1B or Coronin 1B (S2D) was bound to Ni-NTA beads and mixed with 20 nM bovine Arp2/3 complex for 1 hour. Immunoblotting detected Arp2/3 complex bound to the beads.

Figure 4. Interactions between Coronin 1B and Slingshot 1L

A- Rat2 fibroblasts stably expressing Coronin 1B-GFP were treated with 100 nM PMA for 30 min to stimulate Coronin 1B phosphorylation. Cells were washed with fresh media and incubated with the pan-PKC inhibitor Ro32-0432 (1 µM) to inhibit further phosphorylation. Coronin 1B was immunoprecipitated from lysates at the indicated times and blotted with anti-pSer^PKC to monitor dephosphorylation kinetics. B- Dephosphorylation assay carried out as in panel A with or without okadiac acid (100 nM). C- Recombinant Coronin 1B was phosphorylated in vitro by PKC and subjected to in vitro dephosphorylation with SSH1L-myc immunoprecipitated from transfected HEK293 cells. Blots were reacted with anti-pSer^PKC to detect phospho-Coronin 1B. D- Serum starved Rat2 fibroblasts were treated with 100 nM PMA for 30 min and lysates were subjected to dephosphorylation in vitro by increasing concentrations of SSH1L-myc. Phospho-Coronin 1B was detected following immunoprecipitation as described above. Phosphorylation of Paxillin^S126, Erk1/2 and Cofilin were monitored by immunoblotting the unbound fraction after Coronin 1B immunoprecipitation using phospho-specific antibodies. Blots shown are representative of at least three independent experiments. E- Phospho-Coronin 1B was detected in control or HEK293 cells expressing myc-tagged dominant negative mutant SSH1L C393S (CS) as described in panel B. F- Lysates of control Rat2 cells or cells expressing either WT SSH1L-GFP or SSH1L-CS-GFP were blotted with the indicated antibodies. G- Rat2 cells stably expressing either WT SSH1L-GFP or SSH1L-CS-GFP were subjected to the in vivo dephosphorylation assay. Phospho-Coronin 1B was detected by anti-pSer^PKC and quantified by densitometry relative to total Coronin 1B. Results from three independent experiments are presented as means with error bars indicating standard error of the means. H- Lysates from HEK293 cells transiently expressing SSH1L-myc were immunoprecipitated with antibodies to endogenous Coronin 1B or SSH1L-myc and blotted with the indicated antibodies. Immunoprecipitations with rabbit or mouse IgG were performed as negative controls. I- Endogenous Coronin 1B was immunodepleted from lysates of Rat2 cells stably expressing SSH1L-GFP. Control rabbit IgG was used for mock depletion. Residual SSH1L-GFP was immunoprecipitated from the immunodepleted lysates using anti-GFP antibody and blotted with the indicated antibodies. J- Rat2 cells stably expressing SSH1L-GFP were infected with lentivirus expressing Coronin 1B shRNA (KD-1B) or a control shRNA (NS). Arp2/3 complex was immunoprecipitated using anti-p34Arc and blotted for SSH1L using anti-GFP. Similar results were obtained using anti-GFP to immunoprecitate SSH1L-GFP (not shown).

Figure 5. Depletion of Coronin 1B alters lamellipodial dynamics and inhibits SSH1L-induced membrane ruffling

A- Depletion of Coronin 1B inhibits SSH1L-induced hyper-ruffling. Rat2 fibroblasts stably expressing SSH1L or FP4-CAAX were transfected with the Coronin 1B shRNA, and the cells were subjected to kymography analysis as described in Fig. 1E. Newman-Keuls multiple comparison test was used after one-way ANOVA to generate the P values. B- Cofilin S3D suppresses SSH1L-induced hyper-ruffling. Phospho-mimetic mutants of either Coronin 1B (S2D) or Cofilin (S3D) were transiently expressed in Rat2 fibroblasts stably expressing SSH1L-GFP; kymography analysis was carried out as in Fig. 1E.

Figure 6. Depletion of Coronin 1B disrupts targeting of Slingshot 1L to the leading edge

A- Diagram of the lentiviral vector combining Coronin 1B (KD-1B) or NS control shRNA expression from the Pol III U6 promoter and SSH1L-GFP expression from the MSCV 5’ LTR promoter. B- Depletion of Coronin 1B alters the distribution of SSH1L-GFP upon EGF stimulation. Rat2 fibroblasts infected with lentivirus described in panel A were serum starved and stimulated with EGF for 15 min before fixation and staining with anti-Coronin 1B and anti-Cortactin. Regions boxed by yellow dash lines are magnified in the upper panels. Pixel intensities around the cell edge were extracted and plotted as described in Fig. S1. The relative intensities of Cortactin (blue) and SSH1L-GFP (green) are plotted for Coronin 1B depleted cells (KD-1B, upper panel) and for two control cells (NS, lower panels). C- Relative distributions of SSH1L-GFP and Cortactin in control and Coronin 1B-depleted cells. The distance between the maximal intensity of SSH1L-GFP and that of Cortactin staining (panel B) was determined for Coronin 1B-depleted or control cells (n = 27 cells). Unpaired Student’s t-test indicates a significant difference between (KD-1B) and control (NS) cells (P<0.001). D- Cells prepared as described in panel B were stained with anti-Coronin 1B and Alexa Fluor 568-phalloidin. RG-merge shows overlap between SSH1L-GFP and F-actin.

Figure 7. Depletion of Coronin 1B increases phosphorylation of Cofilin and expression of activated Cofilin (S3A) partially rescues the effects of Coronin 1B depletion on lamellipodia dynamics

A- Depletion of Coronin 1B leads to elevated pCofilin, but not via activation of LIMK. Lysates from Rat2 cells infected with lentivirus expressing either the Coronin 1B shRNA (KD-1B) or control shRNA (NS) were blotted with the indicated antibodies. B- Depletion of Coronin 1B decreases the level of active Cofilin in the lamellipodia. Cells generated as described in panel A were immunostained for total Cofilin (MAB22) and phospho-Cofilin (4321) and subjected to ratio imaging. GFP-positive cells in the DIC image express either Coronin 1B (KD-1B) or control (NS) shRNA. Ratios of Active/Total Cofilin were calculated from [(Total Cofilin – pCofilin) / Total Cofilin)] and presented using a rainbow look-up table. C- Active/total Cofilin ratios in whole cells expressing Coronin 1B shRNA (KD-1B) or control shRNA (NS). Unpaired Student’s t-test indicates a significant difference between Coronin 1B-depleted and control cells (P<0.0001). D- Expression of activated Cofilin (S3A) partially rescues the effects of Coronin 1B depletion on protrusive activity. Rat2 fibroblasts infected with Lentivirus for expressing constitutively activated Cofilin (S3A) and Coronin 1B shRNA were subjected to kymography analsysis. Protrusion parameters were determined as described in Fig. 1E. ANOVA test shows significant difference between the samples. Newman-Keuls multiple comparison tests were used to generate P values.